limbxy_seq_t2

This model is a fine-tuned version of c14kevincardenas/beit-large-patch16-384-limb on the c14kevincardenas/beta_caller_284_limbxy_seq_2 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0135
• Rmse: 0.1165

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 2014
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 250
• num_epochs: 15.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rmse
0.0191	1.0	300	0.0182	0.1349
0.0146	2.0	600	0.0158	0.1259
0.0127	3.0	900	0.0135	0.1165
0.0105	4.0	1200	0.0146	0.1209
0.0088	5.0	1500	0.0150	0.1224
0.009	6.0	1800	0.0161	0.1271
0.0071	7.0	2100	0.0166	0.1288
0.0058	8.0	2400	0.0178	0.1336
0.0046	9.0	2700	0.0186	0.1364
0.0045	10.0	3000	0.0190	0.1381
0.0035	11.0	3300	0.0198	0.1408
0.0032	12.0	3600	0.0217	0.1475
0.0026	13.0	3900	0.0218	0.1477
0.0023	14.0	4200	0.0229	0.1514
0.0015	15.0	4500	0.0234	0.1529

Framework versions

• Transformers 4.43.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.19.1
• Tokenizers 0.19.1